SPD Problem of Slow GDT Response

H

Thread Starter

hany5757

It is known that the GDT has a slow response and the TVS is fast response but has low voltage rating.

So my question is: during surge like lightning if very high voltage/current come to the SPD, it will not at first be detected by GDT (due to its low response) and will go to TVS which can not withstand this very high voltage/current and can be burned or damaged. but this actually not happening as the SPD should not be destroyed during surge. So how it works to overcome this issue of slow GDT time response? also please what is actually meant by GDT slow response? is it time taken by it to detect surge or time to response to surge after detecting it?
 
B

Bob Peterson

GDT = gas discharge tube?

SPD are pretty robust. When a high voltage impulse is seen at the SPD it turns on and becomes a very low impedance path, so the voltage across the SPD is not real high for that tiny fraction of a second it is on. The SPD can absorb some level of energy without damaging itself, otherwise it would not be especially useful. Since these transients are very short in duration, even though the current and voltage levels seem high, the amount of energy that the SPD has to turn into heat and dissipate is not all that much.

Here in the US SPD come in several different types. The type used is dependent on where in the distribution chain they are used. Type 1 SPD can be used at the service point and have built in overcurrent protection. Type 2 SPD are intended to be used downstream from the service point and generally require some form of overcurrent protection be provided as part of the installation. Type 3 SPD are used even farther downstream of the service point. Type 1 and 2 SPD generally operate at a higher voltage then type 3, so you end up with this layering effect where a lower level transient might not be acted on by a type 1 SPD.

I would have to look up the exact numbers but I seem to recall that a type 1 SPD designed for 480V delta systems won't operate at all unless the voltage exceeds 2000V L-L (might be more - I don't recall the exact number off hand).

I don't know much about GDT as they are not real common here in the US except as part of some kinds of equipment.

--
Bob
http://ilbob.blogspot.com/
 
> It is known that the GDT has a slow response and the TVS is
> fast response but has low voltage rating.

Sort of true - but there were tritium filled gas tubes that were relatively fast. Most of the other TVS - particularly MOV tend to die in service the first year because they clamp at some voltage - thus P=E^2 * I gets too high and your protection is gone.

>So my question is: during surge like lightning if very high
>voltage/current come to the SPD, it will not at first be
>detected by GDT (due to its low response) and will go to TVS
>which can not withstand this very high voltage/current and
>can be burned or damaged. but this actually not happening as
>the SPD should not be destroyed during surge. So how it
>works to overcome this issue of slow GDT time response? also
>please what is actually meant by GDT slow response? is it
>time taken by it to detect surge or time to response to
>surge after detecting it?

I think you are looking at solving the problem in the wrong way - the key is not to clamp the transient - but to prevent it from getting there in the first place. Back in the early 1980's I designed a transient filter for water systems - They have really bad problems - supper ground - better than the electric utility - Instruments connected to the plumbing - Huge round towers where charges can build on the surface.

Part of the solution was knowing HOW to ground things - which thing etc (A lot of engineers just seem to forget ohms law applies to transients). The other part was realizing crow-bar protection is not a good idea. If you short out a spike - blow a fuse - someone has to replace that fuse - and after several trips they tend to use ever bigger fuses-change out circuit-breakers with bigger ones - they are not happy.

So I used a different method and the installing company was able to guarantee the equipment for 5 years - including lightning hits.

To long to go into here - you can email me at [email protected] if you are still stuck.
 
Top